Can't deploy code to robot

[AaronP]

So basically every time I try deploying code to the robot I get an error that says:

[sshexec] Connecting to roboRIO-1325.local:22
[sshexec] cmd : killall netconsole-host
[sshexec] killall: netconsole-host: no process killed
[sshexec] Remote command failed with exit status 1
[scp] Connecting to roboRIO-1325.local:22
[scp] done.

At the end, it still says "Build Successful". However, when I go into Driverstation it says that I have no Robot Code. I have tried exporting from a different laptop, cycling power on the robot, deploying an older version of the code, and deploying via Ethernet and USB instead of Wifi but none of that works.

Any help would be appreciated. Thanks.

[jajabinx124]

The line "[sshexec] Remote command failed with exit status 1" looks suspicious. Some other teams seem to have encountered the same issue on this thread (http://www.chiefdelphi.com/forums/showthread.php?t=133069).

[iLloyd]

The line "[sshexec] Remote command failed with exit status 1" looks suspicious. Some other teams seem to have encountered the same issue on this thread (http://www.chiefdelphi.com/forums/showthread.php?t=133069).

If I can remember correctly that just means that it could not quit the previous robot code process. This is normal behavior if the Rio was freshly formatted or it doesn't have code.

We had a problem on our team were it would go through the process of uploading the code to the Rio but the DS would not show that Robot Code was available. We found it to be a problem in our RobotInit function that errored out silently. I'd check whatever code is ran when your robot turns on and see if any recent additions could have caused the error. If that doesn't work try to upload a barebones program to see if it works. Lastly try to format your Rio.

[AaronP]

So I tried re-formatting the RoboRIO and that didn't work. However, I tried deploying a barebones program from my laptop and that did not work, but when I deployed it from our classmate I did not get that error and it worked so I have no idea what is going on.

[AustinShalit]

That happens to us when our robot code throws an exception on execution. Have you tried to run your code in debug mode? See the documentation here for info on how to do that: http://wpilib.screenstepslive.com/s/4485/m/13809/l/242588-debugging-a-robot-program

[JohnSmooth42]

We kept having that error for a while and I think it was because I was being silly with pointers(C++).

[Whippet]

We had the same issue, removing any SmartDashboard references fixed it.

[Jared]

The code is deploying just fine. This error simply means that it couldn't find an older running version of your code to kill before loading the new one.

This is very likely because your code is crashing when it starts. Use the riolog to find out what exception is being thrown where so you can fix it.

[cbf]

We had the same issue, removing any SmartDashboard references fixed it.
Were you trying to do an autonomous selector on the SmartDashboard?

Because this code would cause our robot to crash in autonomous:
SmartDashboard::PutData("AutoCommand1Can", autoCommandMoveToZone);
SmartDashboard::PutData("AutoCommand1Can", autoCommand1Can);
SmartDashboard::PutData("AutoCommand1Can", autoCommand1Can1Tote);
The problem was buried deep in the NetworkTables code. I believe one of the STL mapping classes would return a null pointer on an attempt to add a duplicate key to the map, and the code wasn't checking for that.

I've been meaning to report it to the WPILib people, but I've been kind of busy..

[Whippet]

Were you trying to do an autonomous selector on the SmartDashboard?

Because this code would cause our robot to crash in autonomous:
SmartDashboard::PutData("AutoCommand1Can", autoCommandMoveToZone);
SmartDashboard::PutData("AutoCommand1Can", autoCommand1Can);
SmartDashboard::PutData("AutoCommand1Can", autoCommand1Can1Tote);
The problem was buried deep in the NetworkTables code. I believe one of the STL mapping classes would return a null pointer on an attempt to add a duplicate key to the map, and the code wasn't checking for that.

I've been meaning to report it to the WPILib people, but I've been kind of busy..

Yeah, that was the problem.

[MrRoboSteve]

Might have you covered on the bug report.

https://usfirst.collab.net/sf/go/artf4017?nav=1&_pagenum=1&returnUrlKey=1429102010111

[cbf]

Might have you covered on the bug report.

https://usfirst.collab.net/sf/go/artf4017?nav=1&_pagenum=1&returnUrlKey=1429102010111Apparently you do.

I actually had this narrowed to the exact place in Network Tables that this happens, but as this was in the middle of the competition, at the time I was satisfied without simply fixing our own code to make it not happen.

After St Louis, I might give actually go back and and suggest a precise fix.